Two things shape the course of your life: luck and your decisions (with Annie Duke)
Spencer Greenberg and Annie Duke discuss how good decision-making, influenced by probability and magnitude rather than luck, is crucial for individual and societal success. They explore strategies like expected value, avoiding cognitive biases, and the "Monkeys and Pedestals" framework for project management.
Deep Dive Analysis
18 Topic Outline
The Dual Determinants of Life Outcomes: Luck and Decisions
Critique of Traditional Pro/Con Lists in Decision-Making
Improving Decision Processes: Beyond Simple Lists
Understanding and Applying Expected Value Calculations
When to Deviate from Expected Value: Considering Risk of Ruin
Distinguishing Between Aleatory and Epistemic Uncertainty
The Importance of Explicit Decision Calculations
Comparing Intuitive vs. Calculated Decision-Making
The Utility and Limitations of Decision Heuristics
Determining Which Decisions Warrant Deeper Analysis
The Paradox: Why Hard Decisions Are Often Easy
Understanding and Avoiding the Resulting Bias
Applying Non-Resulting Principles in Poker
Evaluating Decisions: Skill vs. Luck in Outcomes
The Omission-Commission Bias and Status Quo Preference
Case Study: The California High-Speed Rail and Quitting Bias
Astro Teller's 'Monkeys and Pedestals' Mental Model
The Mission of The Alliance for Decision Education
10 Key Concepts
Expected Value Calculation
This is a weighted average where each possible outcome's payoff (positive or negative) is multiplied by its probability, and these values are summed. It helps determine the average gain or loss from a decision, guiding choices to maximize long-term benefit for repeated small bets.
Aleatory Uncertainty
This type of uncertainty refers to situations where the probabilities of different outcomes are known, but the specific outcome of a single event is unpredictable. An example is a fair coin flip, where you know it's 50/50 for heads or tails, but not which it will be next.
Epistemic Uncertainty
This uncertainty arises from a lack of knowledge about the situation itself, including unknown probabilities or even not knowing the full scope of possible outcomes. It means there's 'stuff we don't know,' making precise calculations difficult.
Kelly Criterion
A formula for bet sizing that suggests betting a proportion of one's bankroll equal to one's perceived edge in a favorable gamble. It aims to maximize the long-term growth rate of capital while managing the risk of ruin, though often used in a 'half Kelly' or 'quarter Kelly' approach for more safety.
Heuristics
These are mental shortcuts or rules of thumb that allow for faster decision-making. While generally useful and often accurate in common situations, heuristics can lead to systematic errors or biases when applied to edge cases or complex scenarios.
Sorting and Picking (Decision Making)
Decision-making involves two phases: sorting, which is the heavy lifting of identifying and eliminating bad options to narrow down to a set of 'good enough' choices, and picking, which is the faster process of selecting from that refined set of acceptable options.
Resulting
A cognitive bias where the quality of a decision is judged based solely on its outcome, rather than on the information available and the decision process at the time it was made. This can lead to misattributing good outcomes to skill and bad outcomes to poor decisions, hindering learning.
Omission-Commission Bias
This bias describes the tendency to view losses resulting from inaction (omission, e.g., sticking with the status quo) as less severe or blameworthy than losses resulting from an active choice or deviation (commission). It often makes people more tolerant of negative outcomes if they didn't actively cause them.
Sunk Cost Fallacy
This fallacy is the irrational tendency to continue investing resources (time, money, effort) into a project or decision because of past investments, even when current information suggests that continuing is no longer the best course of action. It prevents people from quitting things they've already put a lot into.
Monkeys and Pedestals
A mental model for project planning, suggesting that in highly uncertain projects, one should always tackle the biggest unknown (the 'monkey') first, rather than building easy, known components (the 'pedestal'). This avoids creating an illusion of progress and prevents sunk cost bias from making it harder to quit a failing project.
13 Questions Answered
Improving decision-making is crucial because, alongside luck, the quality of your decisions is one of only two factors determining how your life turns out. Since luck is uncontrollable, focusing on decision quality is the only way to improve outcomes.
Pro/con lists amplify bias because they are 'flat,' meaning they don't distinguish the magnitude or probability of each item, making them easily manipulated unconsciously to favor a desired outcome. They also lack consideration for the likelihood of each pro or con occurring.
While most decisions should aim to maximize expected value, deviations are necessary when considering 'risk of ruin' in situations with high volatility or large bets where a single loss could be catastrophic. Bet sizing strategies like the Kelly criterion can help manage this risk.
The two types are aleatory uncertainty (where probabilities are known, but the specific outcome is random, like a coin flip) and epistemic uncertainty (where there's a lack of knowledge about probabilities or the situation itself, including unknown unknowns).
No, it is always beneficial to try to make expected value calculations explicit, especially for important decisions. Even intuitive decisions involve an implicit forecast, and making it explicit allows for examination, external feedback, and better learning from outcomes.
Intuitive decision-making is essentially a poor or inaccurate expected value calculation, heavily influenced by cognitive biases like pattern matching and recency bias. While it aims to choose the best option, it often contains significant error compared to explicit, deliberate calculations.
Heuristics are useful rules of thumb that speed up decision-making and work well most of the time, similar to how our visual system uses distance cues. However, they can break down in 'edge cases' or novel situations, leading to cognitive biases and errors.
The significance of a decision is determined by its 'impact' (how bad a negative outcome would be, long-term) and 'optionality.' Low-impact decisions with quick feedback loops should be made quickly, while high-impact decisions warrant more time and analysis.
A decision feels 'hard' when there are multiple options that all seem really good. This paradoxically means the decision is actually 'easy,' because the options are so similar in value that you can often just flip a coin without significantly compromising the outcome.
Resulting is judging a decision's quality by its outcome, ignoring the role of luck and available information at the time. It's a big mistake because it causes paralysis from fear of bad outcomes and leads to misallocating future resources by falsely attributing good outcomes to skill or bad outcomes to poor decisions.
The omission-commission bias is our tendency to be more tolerant of losses that result from inaction (omission, e.g., sticking with the status quo) than from an active choice (commission). This bias makes us reluctant to deviate from the status quo, even when a change might lead to better expected outcomes.
The 'Monkeys and Pedestals' model advises tackling the biggest unknown or bottleneck ('monkey') in a project first, rather than starting with easy, known components ('pedestals'). This prevents creating an illusion of progress, avoids sunk cost fallacy, and helps determine project viability early.
The Alliance for Decision Education aims to improve individual lives and society by bringing decision education into every K-12 classroom. They believe teaching decision-making skills, probabilistic thinking, and understanding personal values is crucial for students' future success, more so than many traditional subjects.
20 Actionable Insights
1. Prioritize Decision Quality
Focus on improving the quality of your decisions, as this is the only factor you can control that significantly determines your life’s outcomes, apart from luck.
2. Calculate Expected Value
For important decisions, explicitly calculate the expected value by multiplying the probability of each outcome by its payoff (positive or negative). This makes your forecasts examinable and improves learning from results.
3. Make Forecasts Explicit
Always make your forecasts about future outcomes explicit, rather than relying on implicit gut feelings. This allows for objective examination, external input, and better feedback loops to improve future decisions.
4. Separate Decision from Outcome
Avoid “resulting” by not judging the quality of a decision solely based on its outcome, as luck significantly influences results. This prevents misattributing success to skill or failure to poor decisions, which can skew future resource allocation.
5. Tackle “Monkeys” First
In projects, identify and address the biggest unknowns or bottlenecks (“monkeys”) before investing in known, easier parts (“pedestals”). This prevents wasting resources and makes it easier to quit if the core challenge is intractable.
6. Avoid Low-Hanging Fruit First
Resist the urge to start projects by tackling “low-hanging fruit” (easy wins). Instead, prioritize solving the most critical, uncertain challenges first to avoid creating sunk costs that make quitting difficult later.
7. Overcome Sunk Cost Fallacy
When deciding whether to continue an endeavor, only consider future costs and benefits, ignoring what has already been invested. This helps avoid throwing good resources after bad and enables timely quitting of failing ventures.
8. Recognize Omission-Commission Bias
Be aware of the bias where losses from inaction (omission) are tolerated more than losses from active choices (commission). This can lead to sticking with suboptimal situations to avoid the perceived greater “loss” of making a change.
9. Quantify Status Quo vs. Change
When contemplating a change versus sticking with the status quo, explicitly calculate the expected value of both options. This objective comparison helps overcome omission-commission bias and reveals if a change offers higher expected value.
10. Use Probability & Magnitude
Go beyond simple pros and cons lists by considering the magnitude and probability of both potential gains and costs for each option. This provides a more accurate understanding of the true weight of each factor.
11. Reject Flat Pros/Cons Lists
Do not rely on simple pros and cons lists for important decisions. They amplify cognitive biases by failing to distinguish the magnitude or probability of different factors, leading to potentially manipulated outcomes.
12. Manage Risk of Ruin
For high-stakes decisions or those with significant volatility, consider the risk of ruin and adjust your strategy (e.g., using a fraction of the Kelly criterion). Maximizing expected value alone might lead to catastrophic losses if you can’t withstand the downside.
13. Decide Quickly on Small Bets
For low-stakes decisions with minimal risk of ruin, make choices quickly once a positive expected value is determined. Avoid paralysis by analysis and save your cognitive energy for more impactful decisions.
14. Employ Sorting and Picking
Structure your decision-making into “sorting” (identifying acceptable options) and “picking” (choosing among the acceptable ones). The picking process should always be fast once options are deemed “good enough.”
15. Tailor Decision Time to Impact
Adjust the time and effort spent on evaluating options based on the decision’s impact and optionality. High-impact decisions warrant more thorough sorting, while low-impact ones should be resolved swiftly.
16. Embrace “Hard Decisions are Easy”
When a decision feels hard because you have multiple equally good options, recognize that the options are likely so similar that a quick choice (even a coin flip) is effective. Overthinking small differences is often unproductive.
17. Review Decisions Without Outcomes
When analyzing past decisions, particularly in a group, discuss the information available and the reasoning before revealing the actual outcome. This practice, common in poker, helps avoid hindsight bias and promotes objective learning.
18. Assess Outcome Informativeness
Understand that a single outcome is highly informative only when skill heavily dominates luck (e.g., chess) or when aggregated over many iterations. Avoid drawing strong conclusions from single outcomes in high-luck scenarios.
19. Recognize Heuristic Breakdowns
Be aware that mental shortcuts (heuristics) can break down in “edge cases” or when real-world conditions differ from those in which the heuristic evolved. This helps identify situations where deeper analysis is needed to avoid errors.
20. Advocate for Decision Education
Support and promote the integration of decision-making education into K-12 curricula, teaching concepts like probability, values, and goal-setting. This foundational skill can significantly improve individual lives and society.
5 Key Quotes
When we think about how our lives turn out, there's only two things that determine how our lives turn out. One is luck, and luck we have absolutely no control over. So you'll hear people say, you know, oh, I make my own luck, but you can't actually make your own luck because luck is an exogenous force that just acts upon you. So there's luck. That's one thing that determines how your life turns out. And then the other is the quality of your decisions. And that's it. Those are the two things.
Annie Duke
When a decision's really hard, that actually means it's easy.
Annie Duke
Resulting is not a small mistake. It's a very big mistake because it impacts how you allocate your resources in the future.
Annie Duke
I'm pretty sure that there's almost no dose of LSD high enough that I could beat Magnus Carlsen to chess.
Annie Duke
The low-hanging fruit, by definition, is a pedestal. You should never attack any low-hanging fruit until you've gone to the top of the tree first, until you know that that top of the tree is available to you.
Annie Duke